Deep learning model based on contrast-enhanced ultrasound for predicting early recurrence after thermal ablation of colorectal cancer liver metastasis.

OBJECTIVES: To develop and validate a deep learning (DL) model based on quantitative analysis of contrast-enhanced ultrasound (CEUS) images that predicts early recurrence (ER) after thermal ablation (TA) of colorectal cancer liver metastasis (CRLM). METHODS: Between January 2010 and May 2019, a total of 207 consecutive patients with CRLM with 13,248 slice images at three dynamic phases who received CEUS within 2 weeks before TA were retrospectively enrolled in two centres (153 for the training cohort (TC), 32 for the internal test cohort (ITC), and 22 for the external test cohort (ETC)). Clinical and CEUS data were used to develop and validate the clinical model, DL model, and DL combining with clinical (DL-C) model to predict ER after TA. The performance of these models was compared by the receiver operating characteristic curve (ROC) with the DeLong test. RESULTS: After a median follow-up of 56 months, 49% (99/207) of patients experienced ER. Three key clinical features (preoperative chemotherapy (PC), lymph node metastasis of the primary colorectal cancer (LMPCC), and T stage) were used to develop the clinical model. The DL model yielded better performance than the clinical model in the ETC (AUC: 0.67 for the clinical model, 0.76 for the DL model). The DL-C model significantly outperformed the clinical model and DL model (AUC: 0.78 for the DL-C model in the ETC; both, p < 0.001). CONCLUSIONS: The model based on CEUS can achieve satisfactory prediction and assist physicians during the therapeutic decision-making process in clinical practice. KEY POINTS: • This is an exploratory study in which ablation-related contrast-enhanced ultrasound (CEUS) data from consecutive patients with colorectal cancer liver metastasis (CRLM) were collected simultaneously at multiple institutions. • The deep learning combining with clinical (DL-C) model provided desirable performance for the prediction of early recurrence (ER) after thermal ablation (TA). • The DL-C model based on CEUS provides guidance for TA indication selection and making therapeutic decisions.

Dynamic Contrast-Enhanced Ultrasound Radiomics for Hepatocellular Carcinoma Recurrence Prediction After Thermal Ablation.

PURPOSE: To develop a radiomics model based on dynamic contrast-enhanced ultrasound (CEUS) to predict early and late recurrence in patients with a single HCC lesion ≤ 5 cm in diameter after thermal ablation. PROCEDURES: We enrolled patients who underwent thermal ablation for HCC in our hospital from April 2004 to April 2017. Radiomics based on two branch convolution recurrent network was utilized to analyze preoperative dynamic CEUS image of HCC lesions to establish CEUS model, in comparison to the conventional ultrasound (US), clinical, and combined models. Clinical follow-up of HCC recurrence after ablation were taken as reference standard to evaluate the predicted performance of CEUS model and other models. RESULTS: We finally analyzed 318 patients (training cohort: test cohort = 255:63). The combined model showed better performance for early recurrence than CUES (in training cohort, AUC, 0.89 vs. 0.84, P < 0.001; in test cohort, AUC, 0.84 vs. 0.83, P = 0.272), US (P < 0.001), or clinical model (P < 0.001). For late recurrence prediction, the combined model showed the best performance than the CEUS (C-index, in training cohort, 0.77 vs. 0.76, P = 0.009; in test cohort, 0.77 vs. 0.68, P < 0.001), US (P < 0.001), or clinical model (P < 0.001). CONCLUSIONS: The CEUS model based on dynamic CEUS radiomics performed well in predicting early HCC recurrence after ablation. The combined model combining CEUS, US radiomics, and clinical factors could stratify the high risk of late recurrence.